How machine learning technology could significantly change the role played by life sciences in the future

Artificial intelligence is no longer a promise of the future: it’s making a real difference today in a number of markets. So it’s not really a stretch to imagine that it will have an impact in life sciences.

But where, how and to what extent?

In other industries smart searches that adapt to user preferences, automated personal assistants like Alexa and Siri, and customer care channels such as web chat, are already exploiting AI in everyday situations. Via machine learning, these applications become faster and more accurate over time, adapting to the results they find. All of which helps accelerate and hone decision-making.

The scope for process automation one of the big attractions. As long as humans check the results, why not let IT systems take the load off heavy-duty information processing tasks? They can process masses of data without tiring or missing anything, and can spot the subtlest patterns – at a level humans could never match. Very recently, researchers in Oxford announced AI technology capable of diagnosing heart disease and lung cancer at a much earlier stage, from analysis  of patient scans. Meanwhile continuous patient monitoring is becoming a serious strategy for upholding patient wellbeing instead of passively waiting to fix people. All of which should lead to better patient outcomes.

Redefining life sciences

So where does this leave the life sciences industry, which relies on providing treatments to address and manage existing patient conditions? One possibility is for AI to enhance R&D, through the ability to analyse large volumes of data to produce richer insights – enabling advanced modelling and extrapolation from these findings to drive bolder hypotheses and more targeted research.
Advances in medical imaging interpretation, genomic profiling, personalised medicine and treatments are also within reach, once AI is part of the picture.

AI technology offers a way to track global patient trends too, as well as their concerns, experiences, behaviour and needs. This offers companies new scope to understand more about what is happening in the real world – in turn enabling more proactive and complete monitoring of safety signals as drugs move into markets, plus the chance to spot untapped opportunities.

Preparing for an AI-enabled future

The potential for driving innovation in life sciences is considerable. But certain things must happen first if companies are to exploit the potential ahead of them. First is an acceptance of the need to change as an industry. The second is embracing an IT and data environment that allows for new, controlled experimentation. This isn’t just about developing ‘big data’ strategies, but rather preparing existing data so it can be collated and analysed efficiently using AI platforms.

As things stand, regulatory pressures are the primary driver for most data-related initiatives in this industry. Although more data is being captured, this tends to be for a specific compliance objective, rather than something bolder and more imaginative. Yet it makes sense to keep the bigger picture in focus: if preparations have to be made to meet regulatory targets, isn’t it more economical to build in additional ambitions at the same time, however off the wall some  might seem now?

The good news is that many of the latest regulatory requirements are compatible with AI exploration – even if, for the time being, this is purely in the context of automating routine aspects of submission creation, or content checking, with a view to shortening companies’ time to market.

The critical enabler for all of this (assuming life sciences organisations are open to it) is the creation of an  all-encompassing master data model – one that also allows for interdependencies between data, in a way that can drive new efficiencies and increased impact, once boosted by AI.

Start out with a clear sense of new purpose and see what happens. There is little to lose by being open to new possibilities. Especially if companies are being called to improve their data practices anyway.

Siniša Belina is senior life sciences consultant at AMPLEXOR Life Sciences